Background: Successful deep brain stimulation is mostly dependent on accurate positioning of the leads at the optimal target points. We investigated whether the identification of the subthalamic nucleus in T2-weighted 3-T MRI, fluid-attenuated inversion recovery 3-T MRI and susceptibility-weighted 3-T MRI is confirmed by intraoperative neurological microelectrode recording.

Methods: We evaluated 182 microelectrode recording leads in 21 patients with bilateral deep brain stimulation, retrospectively. Consequently, 728 electrode contact positions in T2-weighted 3-T MRI, 552 electrode contact positions in fluid-attenuated inversion recovery 3-T MRI and 490 electrode contact positions in susceptibility-weighted 3-T MRI were evaluated for a positive nucleus subthalamicus signal.

Results: The highest sensitivity was measured for fluid-attenuated inversion recovery 3-T MRI with 82.5 %, while the highest specificity was observed for susceptibility-weighted 3-T MRI with 90.6 %. The negative predictive value was nearly equal for susceptibility-weighted MRI and fluid-attenuated inversion recovery MRI with 87.5 % vs. 87.1 %, but the positive predictive value was higher in susceptibility-weighted 3-T MRI (86.0 %) than in the other MRI sequences.

Conclusions: The susceptibility-weighted 3-T MRI-based subthalamic nucleus localization shows the best accuracy compared with T2-weighted and fluid-attenuated inversion recovery 3-T MRI. Therefore, the susceptibility-weighted 3-T MRI should be preferred for surgical planning when the operation procedure is performed under general anesthesia without microelectrode recordings.